색상기반 주목연산자를 이용한 정규화된 얼굴요소영역 추출

Normalized Region Extraction of Facial Features by Using Hue-Based Attention Operator

  • 정의정 (LG전자 정보통신사업부 CDMA단말기 연구소) ;
  • 김종화 (경북대학교 전자공학과 병렬분산처리 및 인지과학연구) ;
  • 전준형 (두원공과대학교 컴퓨터) ;
  • 최흥문 (경북대학교 전자공학과 병렬분산처리 및 인지과학연구실)
  • 발행 : 2004.06.01

초록

색상(hue) 기반 주목연산자와 조합누적투영함수(combinational integral projection function: CIPF)를 제안하여 조명변화에 강건하게 정규화된 얼굴요소영역을 추출하였다. 살색 필터를 도입하여 얼굴후보영역들을 추출하고, 거기에 색상과 대칭성에 기반한 주목연산자를 적용하여 조명변화에 강건하게 두 눈의 위치를 정확히 검출할 수 있도록 하였으며, 색상기반 눈 분산 필터로 눈을 검증하여 얼굴영역을 확인하였다. 또한, 색상과 밝기 성분을 조합한 조합누적투영함수를 사용하여 두 눈의 위치를 기준으로 조명변화나 수염의 존재유무에 둔감하게 눈썹 및 입의 수직위치를 구하고, 이를 바탕으로 정규화된 얼굴영역 및 그 요소영역을 추출하였다. AR 얼굴 데이터베이스[8]에 제안한 색상기반 주목연산자를 적용한 결과 기존 명도기반 주목연산자에 비해 약 39.3%의 눈 검출 성능향상을 보임으로써 조명방향 변화에 강건하게 정규화된 얼굴 및 그 요소영역을 일관성 있게 추출할 수 있음을 확인하였다.

A hue-based attention operator and a combinational integral projection function(CIPF) are proposed to extract the normalized regions of face and facial features robustly against illumination variation. The face candidate regions are efficiently detected by using skin color filter, and the eyes are located accurately nil robustly against illumination variation by applying the proposed hue- and symmetry-based attention operator to the face candidate regions. And the faces are confirmed by verifying the eyes with the color-based eye variance filter. The proposed CIPF, which combines the weighted hue and intensity, is applied to detect the accurate vertical locations of the eyebrows and the mouth under illumination variations and the existence of mustache. The global face and its local feature regions are exactly located and normalized based on these accurate geometrical information. Experimental results on the AR face database[8] show that the proposed eye detection method yields better detection rate by about 39.3% than the conventional gray GST-based method. As a result, the normalized facial features can be extracted robustly and consistently based on the exact eye location under illumination variations.

키워드

참고문헌

  1. IEEE Trans. on PAMI v.24 no.1 Detecting Faces in Images: A Survey M.H.Yang;d.J.Kriegman;N.Ahuja https://doi.org/10.1109/34.982883
  2. Pattern Recognition v.27 no.1 Human Face Detection in Complex Background G.Yang;T.S.Huang
  3. Pattern Recognition v.1 Line Extraction and Pattern Detection in a Photograph T.Sakai;M.Nagao;S.Fujibayashi https://doi.org/10.1016/0031-3203(69)90006-5
  4. IEEE Trans. on PAMI v.20 no.1 Neural Network-based Face Detection H.a.Rowley;S.Baluja;T.Kanade https://doi.org/10.1109/34.655647
  5. IEEE Trans. on PAMI v.24 no.5 Face Detection in Color Images R.L.Hsu;M.A.Mottaleb;A.K.Jain https://doi.org/10.1109/34.1000242
  6. IJCV v.14 Context-free Attentional Operators: The Generalized Symmetry Transform D.Reisfeld;H.Wolfson;Y.Yeshurun https://doi.org/10.1007/BF01418978
  7. IEEE Trans. on Signal Processing v.38SP no.1 An Implementation of Noise-Tolerant Context-free Attention Operator and its Application to Efficient Multi-Object Detection C.J.Park;S.H.Cho;H.M.Choi
  8. CVC Technical Reports no.24 The Ar face database A.M.Martinez;R.Benaventa
  9. On-Line Proc. of Vision Interface, Trois-Rivie'res Color edge detection in RGB using jointly Euclidean distance and vector angle S.Wesolkowski;Ed Jernigan
  10. Pattern Recognition v.36 Adaptive skin color modeling using the skin locus for selection training pixels M.Soriano;B.Marinkauppi;S.Huovinen;M.Laaksonen https://doi.org/10.1016/S0031-3203(02)00089-4
  11. Pattern Recognition v.34 multi-cues eye detection on gray intensity image G.C.Feng;P.C.Yuen https://doi.org/10.1016/S0031-3203(00)00042-X
  12. 4th International Conf. on Knowledge-Based Intelligent Engineering Systems & Alied Technologies(KES'2000) v.1 Knowledge-Based Eye Detection for Human Face Recognition L.Zhang;P.Lenders